The need for high performance switching solutions continues to grow in the fields of computing and data handling systems. Examples of such systems include interconnecting computers and high-performance storage devices, interconnecting computers in a multiple-computer operating environment, and anywhere else where multiple high-speed data interconnections must be established between designated nodes or groups of nodes in a data handling network. A switch is a network device at a node that sends and receives data across the network in units of frames. Higher bandwidth and greater switching flexibility are prime concerns for switches and devices to be used in such systems.
The Fibre Channel standard, ANSI X3.T11, is intended to address these concerns. The Fibre Channel standard itself broadly defines classes and standards of performance, but does not dictate the implementation technologies to be used in providing these functions. A particular design of a switch to implement Fibre Channel functions is referred to as the “fabric” of the switch.
In order to increase the physical distance between switches, they often contain optical repeaters that transmit data across the network. The problem is that although the optical repeaters give the distance required, they almost always result in a very low sustainable bandwidth, especially if the link distance between nodes is quite long, such as 100 kilometers or more.
Thus, most conventional switches contain memory called buffers to hold the frames received and sent across the network. Associated with these buffers are credits, which are the number of frames that a buffer can hold per fabric port.
Most existing FC switches have approximately 8-32 credits per fabric port. These easily meet most requirements for longwave and shortwave links. Recently, the demand for longer links has increased, where 100 kilometer links are very popular. 100 kilometer links require approximately 62 credits per link receiver at 1 G, 124 credits at 2 G, and 248 credits at 4 G. It's not always possible, practical or desirable, to have available this much credit at the end of long links, especially FC switches, due to cost and integration concerns. Given especially that switch users would like to connect a long link to any switch port, it forces all switch ports to have a very large credit count or dynamic access to a very large credit count. This is especially impractical given that in larger switch fabrics consisting of multiple switch boxes, the E_Ports (or trunk ports) usually require very little buffering because of the short interconnect. Forcing large credit count buffers onto all ports of a switch increases cost and precludes highly integrated architectures.
In addition, the link has to be routed through optical repeaters on both ends, in order to operate reliably over the 100 kilometers. A typical installation has each end node connected to an optical repeater box, typically in the same room and typically via a shortwave cable. The long link then is actually between the two optical repeaters and what is typically commercially available dark fiber.
Thus, there is a need for a technique to increase the performance of switches.